Optimized timing of using infliximab in perianal fistulizing Crohn's disease.

Infliximab (IFX), as a drug of first-line therapy, can alter the natural progression of Crohn's disease (CD), promote mucosal healing and reduce complications, hospitalizations, and the incidence of surgery. Perianal fistulas are responsible for the refractoriness of CD and represent a more aggressive disease. IFX has been demonstrated as the most effective drug for the treatment of perianal fistulizing CD. Unfortunately, a significant proportion of patients only partially respond to IFX, and optimization of the therapeutic strategy may increase clinical remission. There is a significant association between serum drug concentrations and the rates of fistula healing. Higher IFX levels during induction are associated with a complete fistula response in these patients. Given the apparent relapse of perianal fistulizing CD, maintenance therapy with IFX over a longer period seems to be more beneficial. It appears that patients without deep remission are at an increased risk of relapse after stopping anti-tumor necrosis factor agents. Thus, only patients in prolonged clinical remission should be considered for withdrawal of IFX treatment when biomarker and endoscopic remission is demonstrated, especially when the hyperintense signals of fistulas on T2-weighed images have disappeared on magnetic resonance imaging. Fundamentally, the optimal timing of IFX use is highly individualized and should be determined by a multidisciplinary team. ©The Author(s) 2020. Published by Baishideng Publishing Group Inc. All rights reserved.

Core tip: The long-term outcomes of infliximab in the treatment of perianal fistulizing Crohn’s disease are unfavorable, due to loss of response. The optimization of the therapeutic strategy may increase clinical remission. Higher infliximab concentrations during induction are associated with a complete fistula response. Only patients in prolonged clinical remission should be considered for withdrawal of infliximab when biomarker, endoscopic and radiological remission is demonstrated. Fundamentally, the optimal timing of infliximab use is highly individualized and should be determined by a multidisciplinary team.

INTRODUCTION

Crohn's disease (CD) is a chronic, disabling and idiopathic inflammatory bowel disease that can involve any element of the gastrointestinal tract. Perianal fistulas are a common extraintestinal manifestation of CD, presenting in approximately 40% of patients before or at the time of diagnosis and in 24% after diagnosis[1]. The median interval between CD diagnosis and the first perianal fistula is 2.9-4.4 years[1,2]. Perianal fistulas correlated with CD are indicative of severe disease and a more aggressive course. The natural progression of perianal fistulizing CD (PFCD) is characterized by alternation of remission and relapse periods. The recurrence rate is up to 80% after a median follow-up of 10 years[2]. Repeated perianal symptoms, such as persistent purulent discharge, pain, and fecal incontinence, can cause fatigue, anxiety, and depression, which can be debilitating and negatively impact patients’ quality of life. As the disease progresses, fecal diversion may be necessary to achieve clinical remission in the advanced period. However, it is a fatal procedure because the success rate of restoring bowel continuity is only 16.6%[3]. Ultimately, proctectomy is performed in 41.6% of patients suffering from fecal diversion failure[3].

Emerging biologic agents have revolutionized the medical treatment of PFCD and achieved more promising outcomes than immunomodulators[4]. In the biological era, the treatment goal has changed from symptom relief to complete fistula healing, while also preventing relapse. Fistulizing CD was, together with steroid dependence or resistance, the first indication for biological therapy, after surgical drainage of any sepsis[5]. Infliximab (IFX) is the first anti-tumor necrosis factor (TNF) agent for the treatment of PFCD. The ACCENT trial showed that 68% of patients with fistulizing CD achieved symptom improvement following IFX monotherapy, whereas the closure rate of fistulas was only 36% at 54 wk[6,7]. This finding indicated that a substantial proportion of patients partially responded to IFX. Surgical interventions appear to be indispensable in assisting IFX to alter the natural course of PFCD, because the probability of perianal surgery does not significantly decrease after the emergence of biologic agents[8]. Anatomically, CD-related perianal fistulas can be categorized into two types: Simple and complex[9]. Fistulotomy achieves excellent outcomes in the treatment of symptomatic simple fistulas, with a recurrence rate of 3.4% during a mean follow-up of 1.6 years[10]. However, complex fistulas that are associated with an increased risk of fecal incontinence make up a larger proportion in PFCD. Although sphincter-preserving procedures, such as loose-seton and ligation of the intersphincteric fistula tract (LIFT), show promising outcomes in the treatment of PFCD, they might be restricted by concomitant proctitis in the early stage of the disease[11,12]. The optimal timing of IFX combined with perianal surgery is unclear due to a debate on the relationship between proctitis and surgical outcomes. Poor prognosis obliges the optimization of IFX therapy to induce a more complete response, alter the natural course of PFCD, and reduce complications, hospitalizations, and the incidence of major abdominal surgery.

To date, there is absence of a consensus on the optimal timing of IFX use. The purpose of this review is to examine the present state of knowledge regarding the use of IFX in PFCD patients in order to contribute to the better management of PFCD.

WHEN TO INITIATE THE PRESCRIPTION OF IFX

Owing to the clinical application of biologics, the healing rate of PFCD has improved. The capability of anti-TNF agents to modify the natural course of PFCD has been validated. The cumulative incidence of PFCD at 10 years has decreased from 24% in the prebiologic era to 12% in the biologic era; similarly the probability of proctectomy has decreased from 24% to 13%[1]. An increasing proportion of CD patients switch to biologics. Although IFX has been recommended as the first-line therapy for PFCD by current European Crohn's and Colitis Organisation consensus, there is still a divergence in the “top-down” strategy due to the hidden perils of overtreatment and severe adverse events[13].

Colombel et al[14] demonstrated that patients treated with IFX alone showed a higher healing rate of intestinal mucosa than those treated with azathioprine monotherapy. A real-life study showed that IFX as the first-line therapy was mainly applied in patients with risk factors, higher disease activity and lower quality of life scores[15]. PFCD patients who have a poorer prognosis may benefit from the early introduction of IFX. IFX immediately works to treat PFCD after its first infusion, while the effect-beginning time of adalimumab is over 4 wk after injection[16]. Moreover, the response rate of fistulizing CD to IFX is negatively related to the disease duration[17]. The “step-up” approach may potentially increase the loss of response due to a prolonged disease course and disease progression. Conversely, IFX used as the initial medication can rapidly induce clinical remission and prevent disease progression. Regarding adverse events, infection is the most common, accounting for 53.7% of CD patients treated with IFX[18]. However, the incidence rate of serious infections is only 2.15%[18]. Mortality and malignancy rates are similar between IFX-treated patients and patients with other treatments. Nonserious cerebrovascular accidents and pulmonary embolisms occur in less than 0.1% of the IFX-treated patients[18]. In light of the above evidence, a “top-down” strategy is better for the treatment of PFCD.

TIMING TO COMBINE IFX WITH DEFINITIVE SURGERY

It is well known that surgical intervention is necessary for the drainage of septic complications before the initiation of IFX therapy. However, whether definitive surgery is needed is controversial since IFX can induce fistula closure in approximately 60% of PFCD patients[19]. Despite clinical closure, most fistula tracts can be visualized on pelvic magnetic resonance imaging (MRI). Perianal surgery can improve fistula response to IFX and promote deep remission. It is reported that IFX combined with surgery can improve clinical efficacy compared to monotherapy[20].

Active proctitis negatively affects the outcomes of PFCD, which determines the timing of IFX combined with surgery. Conventionally, surgical procedures are only considered after the elimination of proctitis by prior IFX therapy[21]. In a small sample size study, definitive surgeries, such as fistulotomy and advancement flap, were performed after proctitis was well controlled, with a median interval of 9 wk between the first infusion of IFX and surgery [22]. The healing rate of perianal fistulas was 80% with a median follow-up of 17.5 mo. Nonetheless, the failure of fistula closure may increase in patients with a partial response or primary nonresponse to IFX due to the increased aggression and complexity of perianal fistulas.

The authors performed loose-seton with the eradication of the internal opening within 1 wk before the first infusion of IFX. The clinical healing rate of perianal fistulas was 96.4% after a median follow-up of 26.4 mo[11]. Another study showed that proctitis was detected in 62.5% of patients who achieved improvement of PFCD following definitive surgery[23]. In a prospective study including 15 patients with PFCD, the healing rate of perianal fistulas following LIFT was 67%, with a follow-up duration of 12 mo and without fecal incontinence (Figure 1)[12]. This is comparable with the success rates in cryptoglandular anal fistulas[24,25]. In the cohort, 9 patients had active proctitis, but this finding was not closely related to LIFT failure. Pretreatment with biologic therapy did not improve the outcomes of LIFT[12,26]. A recent multicenter retrospective study demonstrated that multimodal treatment at the diagnosis of PFCD could reduce the probability of repeat surgery and proctectomy[27].

Figure 1

Ligation procedure of the intersphincteric fistula tract for Crohn’s disease-related perianal fistula. A: Identification of the fistula tract with a probe; B: Ligation of the intersphincteric tract; and C: Suture of the intersphincteric incision following curetting the remnant tract.

In addition, the issue of wound healing can be addressed by amelioration of immune inflammation, as the median response time of PFCD to IFX is only 9 d[28]. Early combination therapy without waiting for the disappearance of proctitis is viable and is of great importance, as it can alter the natural course of PFCD as soon as possible and improve the patients’ quality of life.

TIMING TO MONITOR AND OPTIMIZE IFX THERAPY

Perianal lesions predict an increased risk of loss of response. Better outcomes are associated with response monitoring and the timely optimization of the therapy regimen during the induction and maintenance of IFX. Nevertheless, the coexistence of luminal and perianal diseases makes the process monitoring and optimization complex and difficult.

Pelvic MRI

It is inadequate to assess the response of PFCD according to clinical symptoms, especially in terms of discriminating between a closed and healed fistula. The inaccurate assessment of the fistula healing process might misguide the adjustment of the IFX therapy regimen, resulting in worse therapeutic efficacy. Pelvic MRI has been suggested as the gold standard for the assessment of the anatomy and activity of CD-related perianal fistulas. Fistula healing is characterized by the disappearance of hyperintense signals on T2-weighted images and the absence of contrast enhancement after gadolinium injection on T1-weighted fat-suppression images (Figure 2)[29]. After anti-TNF therapy, healed fistulas confirmed by MRI account for 50%-61.5% of closed fistulas[30-32]. Persistent tracts indicate a large probability of recurrence and a prolonged duration of maintenance therapy.

Figure 2

Deep remission of Crohn’s disease-related perianal fistula on magnetic resonance imaging. A: Hyperintense signal on T2-weighted fat-suppression imaging showing an active suprasphincteric fistula; B: Disappearance of hyperintense signal on T2-weighted fat-suppression imaging displaying deep remission of the fistula.

Proctitis increases the risk for PFCD occurrence and recurrence. The formation of perianal fistulas occurs in 92% of CD patients with rectal involvement[33]. The absence or disappearance of rectal involvement plays a pivotal role in the deep remission of PFCD, which is defined as clinical remission associated with absence of anal canal ulcers and healing on MRI[31]. In the majority of studies, thickening of the rectal wall was considered an indicator of proctitis[31,32]. In addition, a recent study demonstrated that the size of the mesorectal lymph nodes, mural fat and creeping fat were also relevant to the evaluation of proctitis by pelvic MRI[34].

Changes in the signal intensity and morphology of fistulas and the rectum can indicate the healing or worsening of PFCD. Scheduled pelvic MRI examinations can provide objective evidence for the assessment of treatment efficacy and the optimization or modification of the therapeutic strategy. The monitoring timing varies. The probability of clinical remission is 5 times greater in PFCD patients with a clinical response to anti-TNF agents within 6 wk than in those responding longer than 6 wk[35]. Hence, the sixth week within induction period is a critical time point to evaluate the response of perianal fistulas and proctitis by pelvic MRI. Features of proctitis on MRI are significantly correlated with those on endoscopy during the maintenance therapy period[34]. Given that pelvic MRI is noninvasive and does not have radiation risk, it could be used to dynamically monitor the healing of PFCD at intervals of 8 wk, acquiring accurate information and providing personalized treatment. Radiological healing might lag behind clinical remission by 12 months, suggesting that MRI monitoring should be carried out for at least 1 year[35,36]. Prolonged treatment is often needed to observe the eradication of fistula tracts on MRI.

Trough levels of IFX

When the loss of response is observed on pelvic MRI, clinicians should check the serum trough levels of IFX. The exact mechanism of the loss of response is unclear, but may be associated with drug metabolism or the formation of antidrug antibodies. After exposure, specific antibodies secreted by clonally expanded lymphocytes form immune complexes with IFX. This process is also termed immunogenicity and may cause excessive drug clearance via the reticuloendothelial system. The levels of antibodies to IFX have been shown to be higher in patients with a loss of response than in those who maintained remission[37].

Increasing evidence suggests that low serum trough IFX levels are related to a lack or loss of response[38]. Although a cut-off level of 5.0 μg/mL is recommended as the target concentration for healing the intestinal mucosa, a specific level related to the complete response of PFCD has not been identified[39]. In a recent retrospective cross-sectional study including 29 PFCD patients receiving IFX, higher than 7.1 μg/mL was identified as the optimal threshold value for fistula healing (77.8% sensitivity and 100% specificity)[40]. The median trough concentrations in patients with healed fistulas were significantly higher than those without healed fistulas (8.1 μg/mL vs 3.2 μg/mL). Fistula healing was positively related with trough IFX levels. Another similar study with a larger sample size indicated that trough IFX levels above 10.1 μg/mL at 4 wk might provide better outcomes for PFCD[41]. Davidov et al[42] demonstrated that trough IFX levels of 9.25 μg/mL at week 2 (89% sensitivity and 90% specificity) and 7.25 μg/mL at week 6 (80% sensitivity and 83% specificity) were the best response predictors of perianal CD. The inconsistency of outcomes may be caused by the various assays and different testing time. Further studies are required to determine the optimal measurement time of drug concentrations and the target IFX levels for fistula healing. More attention should be paid in the induction phase, where multiple factors, such as tissue IFX levels, low albumin, and protein loss, affect the serum drug concentrations.

Therapeutic regimen optimization

As mentioned above, adequate drug concentration is a crucial part of a treat-to-target strategy. The aim of therapeutic regimen optimization is to achieve a steady-state range of serum drug concentrations. Since a higher trough IFX level is necessary for fistula healing than that for mucosal healing, dose escalation should be primarily considered for PFCD patients who do not achieve a response or deep remission prior to switching therapy. Additionally, low drug concentrations can stimulate the germination of immunogenicity, which may be mitigated by early dose optimization. Preexisting antidrug antibodies may be spontaneously degraded in a portion of patients with the continuation of IFX treatment, which also supports the consideration of dose escalation following a loss of response[43]. A dose increase and/or a reduction in the infusion interval are mainly used for increasing serum IFX levels. After dose escalation, 84.8% and 62.3% of CD patients achieved a response, respectively, during the induction and maintenance periods[44]. In terms of safety, trough IFX levels above 7 μg/mL can provide better outcomes for CD patients without increasing the risk of infection[45].

At 54 wk after IFX treatment, antidrug antibodies that were responsible for a loss of response are detected in 62.1% of CD patients[46]. IFX combined with azathioprine is recommended to reduce immunogenicity and mitigate the development of antidrug antibodies. Concomitant therapy can increase serum trough levels of IFX and prolong the duration of fistula closure in CD patients[47,48]. However, early immunosuppressive administration has no effect in increasing clinical remission[49,50]. Furthermore, concomitant therapy does not show better efficacy than IFX monotherapy among CD patients with similar serum IFX levels[51]. Optimized IFX monotherapy leads to similar clinical efficacy as combination therapy[52]. As dose escalation is limited by the increased risk of serious adverse events and increases the consumption of IFX, azathioprine as an adjunct plays a role of dose-sparing by improving the pha-rmacokinetic features of IFX.

The positive rates of antibodies to IFX were 1.6% at 2 wk, 3.3% at 6 wk, and 17.2% at 14 wk[46]. This discrepancy suggests that a drug concentration below 7 μg/mL at 14 wk is an independent predictive factor for long-term nonresponse. Hence, dose escalation or the addition of immunomodulators within 14 wk can increase the clinical response and remission by elevating serum IFX levels. In addition, the benefits of concomitant therapy should be weighed against the increased risk of serious and opportunistic infections[53].

After IFX failure, it may be beneficial to switch to other biologic agents. Adalimumab (ADA) is another effective anti-TNF agent for the treatment of PFCD, which can maintain remission in 41% of patients naïve to anti-TNF drugs at 12 mo[54]. Moreover, ADA, as a second-line therapy, induced complete response in 50% of PFCD patients refractory to IFX[55]. Previous administration of IFX does not affect the efficacy of ADA induction of fistula closure[56]. Although certolizumab pegol, vedolizumab, and ustekinumab show potential benefits for PFCD patients who failed in IFX or ADA therapy, the dedicated efficacy needs further investigation with large sample size studies[57-59].

TIMING TO WITHDRAW IFX

IFX withdrawal is an important question faced by patients and clinicians after disease remission, due to safety and cost-effectiveness concerns. It is well known that the cessation of IFX therapy after sustained clinical remission is responsible for the recurrence of CD. It has been shown that 29.4%-49.3% of patients with remission experienced relapse within 1-4 years after stopping anti-TNF therapy[60-62]. Overall, approximately 20% of patients never received retreatment with a biologic within a long-term follow-up[63,64]. Fortunately, clinical response can be successfully induced by retreatment with the same anti-TNF agents, primarily IFX, in 80%-94% of cases[60-62]. The high rate of secondary remission may counterbalance the high rate of relapse after withdrawal, suggesting that the discontinuation of IFX therapy and the establishment of a cyclic therapeutic strategy consisting of drug discontinuation and retreatment may be possible[65].

Currently, the decision to withdraw IFX treatment is based on the guidelines for luminal CD because of the absence of dedicated guidelines for PFCD[66]. Heterogeneity of disease phenotype and the absence of controlled trials make it difficult to draw decisive conclusions. Deep remission, defined as clinical remission associated with endoscopic and radiological remission, seems to be the criterion for IFX withdrawal. However, the outcomes are unfavorable, with a relapse rate of approximately 55%[35,63]. The risk factors for relapse after withdrawal included ileal localization at diagnosis, a persistent external opening, prior dose optimization, anemia and a white blood cell count above 5 × 109/L at the time of withdrawal[63,64]. Despite the elimination of risk factors, the optimal timing for withdrawal after deep remission is still unknown, which may affect disease progression. Given that after withdrawal, the relapse of disease is apparent while the clinical benefits, such as a reduction in infection or cancer risk, are theoretical because of the absence of controlled studies, maintenance IFX therapy over a longer period may be more beneficial for PFCD patients. IFX discontinuation as a part of a cyclic therapeutic strategy may be implemented in strictly selected patients. The definitive interruption time should be clarified in future studies.

CONCLUSION

In general, no single treatment can successfully manage PFCD. Although IFX has been recommended as a first-line therapy, early combination with definitive surgery may rapidly lead to clinical remission. Monitoring drug concentrations plays a pivotal role in the optimization of the therapeutic regimen. Scheduled MRI scans can dynamically monitor remission of the internal tract in order to immediately adjust the treatment strategy (Figure 3). IFX withdrawal seems to be possible in the setting of deep remission but is not recommended. The optimal timing of IFX use is highly individualized and should be determined by a multidisciplinary team composed of gastroenterologists, colorectal surgeons, and radiologists.

Figure 3

Therapeutic strategy of perianal fistulizing Crohn’s disease. MRI: Magnetic resonance imaging.